[Info-vax] OpenVMS Development Annoyances

[IanD]

ACMS was good for lots of things but having to chase down a process or task death in the ACMS logs could get tedious. It also got hairy at times, especially if it crashed at a lower level and you then had to wade through swlup logs when ACMS didn't start for example

I used to have to do a lot of work with dectrace to delve into ACMS tasks for performance issues. I enjoyed that a lot (not being sarcastic either)

We ran with 7 vaxes as online nodes feeding two alpha gs1280 rdb database nodes. When everything was in a cluster using dectrace to span across ACMS and rdb, things were good.

Eventually we split up the cluster because the vaxes slowed us down too much, after that, dectrace with ACMS and rdb became a lot more painful to debug.

I've seen instances where a machine crashed inside an ACMS task and hp could not determine the cause even after we gave them full crash dumps. It seemed as though dectrace, ACMS and rdb knowledge back then was waning, especially when dectrace and rdb went to Oracle, the cross knowledge understanding went out the door

The other pain was getting hold of the dectrace definitions for rdb. ACMS wasn't too bad from memory but the whole loading etc and having to stop the dectrace database was annoying.
It was a while ago now, so my memory is patchy but exactly matching rdb and dectrace definitions was often a bit of an afterthought for dectrace it seemed

I used to enjoy pulling apart an ACMS task and seeing where the bottlenecks were, then again, I've always loved statistics

I used to plead with programmers to add dectrace calls in their code but it often feel on deaf ears. Dectrace can be very helpful
.
Thankfully ACMS and rdb have these added automatically and you can narrow down what part is the procedure call and task set up times and what not (I'm struggling to remember the components of an ACMS task in dectrace now *sigh")

I even tried to use an rdb collection and feed it into rdbexpert. It did ok although the last collection I took exceeded something with rdbexpert and it crashed and even though I logged a ticket with hp, they basically didn't/couldn't do anything with the limitation and just told me to make my collections smaller. The volume of data collected was high so that really wasn't a workaround. I suspect this was around the time dectrace, rdb and rdbexpert might have been heading off for Oracle land? I don't exactly remember now

I was always impressed with ACMS, rdb and dectrace as products and how they functioned together well

It seemed to stem from a like minded design philosophy, unlike the modular crap on sees in systems and products today where it's blatantly obvious different teams design and write different parts. They often cannot even get a definition sentence consistent